Sulfur retention and particle motion during fluidized bed combustion of coal

Sulfur Retention

and

Particle Motion

during

Fluidized Bed Combustion of Coal

OFF-GAS ( S 0

)

AIR

COAL

SO

ASH

SPENT

SORBENT

schouten

Sulfur Retention

and

Particle M o t i o n

during

Fluidized Bed Combustion of Coal

Copyright © 1988 by J.C. Schouten, Delft, The Netherlands.

No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form, or by any means, electronic, mechanical photocopying, recording or otherwise, without the written prior permission of the author.

Sulfur Retention

and

Particle Motion

during

Fluidized Bed Combustion of Coal

PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Technische Universiteit Delft,

op gezag van de Rector Magnificus, prof.dr. J . M . Dirken, in het openbaar te verdedigen ten overstaan van een commissie

door het College van Dekanen daartoe aangewezen, op donderdag 24 maart 1988

te 14.00 uur door

JACOB CORNELIS SCHOUTEN scheikundig ingenieur

Delft University Press / 1988

TR diss

1619

Dit proefschrift is goedgekeurd door de promotor prof.drs. P.J. van den Berg en door de toegevoegd promotor ir. C.M. van den Bleek.

De promotiecommissie wordt gevormd door: voorzitter:

De Rector Magnificus der Technische Universiteit Delft of diens vervanger. promotor:

prof.drs. P.J. van den Berg ITechnische Universiteit Delft, Faculteit der Scheikundige Technologie en der Materiaalkunde)

toegevoegd promotor:

ir. C.M. van den Bleek (Technische Universiteit Delft, Faculteit der Scheikundige Technologie en der Materiaalkunde)

leden:

prof. B. Scarlett, M.Sc. (Technische Universiteit Delft, Faculteit der Scheikundige Technologie en der Materiaalkunde)

prof. dipl.-ing. K.R.G. Hein (Technische Universiteit Delft, Faculteit der Werktuigbouwkunde)

prof.dr.ir. J . J . H . Brouwers (Universiteit Twente, Faculteit der Werktuigbouwkunde) dr.ir. H.A. Masson (Université Libre Bruxelles, Mécanique Appliqué,

Brussel, België)

Ir. C.M. van den Bleek heeft als begeleider in hoge mate bijgedragen aan het

totstandkomen van het proefschrift. Het College van Dekanen heeft hem als zodanig aangewezen.

STELLINGEN

behorende bij het proefschrift ter verkrijging van de graad van doctor in de technische wetenschappen SULFUR RETENTION and PARTICLE MOTION during FLUIDIZED BED COMBUSTION of COAL

van Jaap Schouten Delft, 24 maart 1988

Er bestaat een paradox tussen enerzijds de veelzijdige en complexe onderlinge samenhang van de fysische verschijnselen en chemische processen in een wervelbedreaktor en anderzijds de betrekkelijke eenvoud van de modellen waarmee deze verschijnselen en processen (veelal) onafhankelijk van elkaar worden beschreven.

Een sluggend fluide bed bestaat uit twee duidelijk onderscheidbare delen met verschillende aanwezigheidswaarschijnlijkheden van de vaste stof. Het beddeel boven de bedhoogte bij minimale fluldisatie kan beschouwd worden als een grote 'splash zone' waar de aanwezigheidswaarschijnlijkheid in axiale richting sterk afneemt t.g.v. een verdunningseffect [1]. Bij de keuze van de plaats van continue aftap van sorbentmateriaal in een FBC-installatie moet rekening worden gehouden met dit fluid dynamisch gedrag van het bed. Wanneer de vaste stof onderhevig is aan segregatie, dient de sorbentaftap gekozen te worden beneden of boven de bedhoogte bij minimale fluldisatie afhankelijk van de aard van de segregatie van het sorbentmateriaal ('jetsam' of 'flotsam').

[1] Dit proefschrift.

De enige juiste manier voor het vaststellen van de wijze van gasdoorstroming en gasoverdracht tussen fasen In een fluide bed is het experimenteel bepalen van de axiale en radiale concentratieprofielen in zowel de bellen als de dichte fase in het bed [1). Het toepassen van theoretische reactormodellen, met behulp waarvan de discriminatie tussen verschillende gasdoorstromings-mogelijkheden (zoals 'ISTR' of '(dispersed) plug flow'; 'fast bubble' of

'slow bubble') geschiedt op grond van alleen de concentratie aan de uitgang van de reactor [2], moet worden afgewezen.

[1] Almstedt, A.-E. and Ljungstrom, E.B., Proc. 9th Int. Conf. on FBC. volume 1, 575-585, Boston, USA, May 3-7, 1987.

12] Van der Looij, J.M.P., Proc. Int. Svmp. on Multiphase Flows. ISMF 1987.

volume 1, 19-27, Zhejlang University, Hangzhou, People's Republic of China, August 3-5, 1987.

-Voor de uitvoering van representatieve zwavelretentie-experimenten tijdens stationaire bedrijfstoestand in een FBC-installatie vormt het een essentiële voorwaarde dat een meting niet eerder wordt uitgevoerd dan na een periode van ten minste 5 maal de gemiddelde sorbentverblijftijd. Als criterium voor het bereiken van de stationaire toestand moet daarbij niet gehanteerd worden het constant zijn van bijv. de gemiddelde bedtemperatuur en/of gassamenstelling (zoals veelvuldig in de praktijk wordt toegepast), maar veeleer het constant zijn van de gemiddelde samenstelling van het sorbentmateriaal in het bed. Van den Bleek, C.M. and Schouten, J.C., Written comment on Proc. 9th Int. Conf. on FBC. Boston, USA, May 3-7, 1987 (to be published by the Amer. Soc. Mech. Eng. (ASME)); Delft, June 3rd, 1987.

In de zwavelretentiemodellering is het gebruik van eenvoudige (empirische) kinetiekrelaties voor de sulfatatiereactie essentieel voor het verkrijgen van een eenvoudig en inzichtelijk model ([1] en [2]). Wanneer deze relaties echter niet correct worden toegepast in zowel de vaste-fase balans als de gas-fase balans, worden weliswaar eenvoudige doch incorrecte vergelijkingen verkregen voor de zwaveldoorbraakconcentratie en de molaire calcium/zwavel-verhouding in de reactorvoeding ([3] en [4]).

[1] Schouten, J.C. and Van den Bleek, C.M., Shorter Communication submitted for publication to Chem. Eng. Sci., 1986.

[2] Schouten, J.C. and Van den Bleek, C.M., Proc. 9th Int. Conf. on FBC. volume 2, 749-761, Boston, USA, May 3-7, 1987.

[3] Zheng, J., Yates, J.G. and Rowe, P.N.. Chem. Eng. Sci.. 22 (2), 167, 1982.

[4] Noordergraaf, I.W., PhD Thesis. Department of Chemical Engineering and Chemistry, Delft University of Technology, The Netherlands, 1985.

Bij het ontwerpen van een fluide-bed-reaktor verdient het aanbeveling om de toepassing van een 'konisch' gevormde reaktor te onderzoeken. In een derge­ lijke reaktor worden de verschillende verschijnselen en processen in het bed simultaan beïnvloed door de gelijktijdige verandering in axiale richting van de beddiameter, de H/D-verhouding en de superficiële gassnelheid.

In de wervelbedverbranding van steenkool biedt een konische reactor de vol­ gende voordelen: verbetering van de zwavelretentie door vergroting van de gasverblijftijd [1]; afname van segregatie t.g.v. de toenemende beddiameter [2]; vergroting van het koolverbrandingsrendement door vermindering van fines-elutrlatie; afname van NOx-vorming d.m.v. getrapte verbranding in een 'sandwiched B-A' fluïde bed [3].

Een 'serieschakeling' van reaktordelen met verschillende diameters biedt In dit kader eveneens aantrekkelijke mogelijkheden.

[1] Ulerich, N.H., Newby, R.A. and Keairns, D.L.. Report EPRI CS-3330. Project 1336-1. 5-6/5-7, Westinghouse Electric Corporation, Research and Development Center, Pittsburgh (PA), USA, December 1983.

[2] Dit proefschrift.

[3] Masson, H.A. and Stievenart, Ph., Lezing gepresenteerd op WUB werkgroep bijeenkomst, Universiteit Twente, Enschede, 18 september 1987.

7. De mate van zwavelretentie in een fluide-bed-verbrandingsinstallatie kan niet voortdurend toenemen door vergroting van de hoeveelheid aktief sorbent materiaal in het bed. Dit houdt in dat bij een bepaalde maximale CaO-omzetting niet altijd een beduidende toename van de zwavelretentie bij een constante (Ca/S)-verhouding kan worden verkregen door het gebruik van een ander sorbent materiaal met een grotere maximale CaO conversie.

Dit proefschrift.

8. De beweging van de vaste fase in een fluïde bed wordt meestal beschreven met het verschijnsel 'dispersie', dat gebaseerd is op de willekeurige beweging van de deeltjes (bijv. het één-dimensionale dispersie model van Taylor [1]). Deze aanpak is vergelijkbaar met de diffusie van molekulen in een gas, welke gerelateerd is aan een concentratiegradiënt. Op dezelfde wijze wordt de dispersie van deeltjes gerelateerd aan een verschil in deeltjesconcentratie in het bed. Een aannemelijke fysische verklaring voor deze beschrijving van het deeltjestransport is niet voorhanden. Het is beter de beweging van vaste stof in een fluïde bed te beschrijven met een twee-fasen model waarbij er deeltjesuitwisseling plaats vindt tussen omhoog- en omlaaggerichte convec-tieve stromen van vaste stof [2],

[1] Avidan, A. and Yerushalmi, J., AIChE J., 31 (5), 835, 1985.

[2) Schouten, J.C., Masson, H.A. and Van den Bleek, C.M., Proc. Int. Svmp. on Multiphase Flows. ISHF 1987. volume 1, 237-243, Zhejiang University, Hangzhou, People's Republic of China, August 3-5, 1987.

9. Bij de bestudering van de chemische en/of intrinsieke kinetiek van een batch gas-vast-reactie op laboratoriumschaal is de keuze van de reactor van wezenlijk belang. In het algemeen geldt de regel dat deze reactie ook moet worden bestudeerd in het type reactor waarin op praktijkschaal het proces wordt uitgevoerd. De aard van de gas-vast-menging in deze reactor bepaalt niet de grootte van de chemische/intrinsieke reactiesnelheid (want kinetiek en hydrodynamica vormen uiteraard onafhankelijke systeemeigenschappen), maar kan wel van invloed zijn op de grootte van de conversie van verschillende vaste-stof-deeltjes op hetzelfde tijdstip in de reactor en daarmee op de vereiste complexiteit van de te hanteren kinetische relaties.

Dit proefschrift.

10. De algemene toepasbaarheid van een mathematisch fysisch/chemisch model ten behoeve van wetenschappelijke en/of industriële (proces)technologische toepassing wordt bepaald door drie factoren welke, in volgorde van belang­ rijkheid, gegeven zijn als:

1. de eenvoud van het gebruik van het model; zowel ten aanzien van de berekening van de fysisch/chemische modelparameters, als ten aanzien van de vereiste (numerieke) berekeningsmethode;

2. de mogelijkheid tot experimentele toetsing en nauwkeurige kwantitatieve validatie van het model;

3. de mogelijkheid tot en de betrouwbaarheid van de modelmatige opschaling van laboratoriumgegevens naar pilot-plant of grootschalige toepassingen.

-11. De overtuigingskracht welke spreekt uit een politieke redevoering is voor­ namelijk gelegen in een afgewogen toepassing van de uitspraak van Einstein dat "everything should be made as simple as possible, but not simpler".

12. Corrosiebestrijding in de agrarische sector vraagt om aktie; rust roest ook in de landbouw!

Schouten, J.C., Landbouwmechanisatie, 34 (5), 551-553, 1983.

Schouten, J.C. en Gellings, P.J., TH&MA, Extern kwartaalblad THTwente, 9, 18-21, Juli 1983.

13. Bedrijven en instellingen die thans niet deelnemen aan de trend om meer vrouwen aan te trekken voor leidinggevende funkties, zullen daarvan in de toekomst hinder ondervinden in hun relaties met partners, concurrenten en klanten.

IA. De instelling van een Corrosie Informatie Transfer Punt ten behoeve van de agrarische sector is zeer gewenst.

Schouten, J.C., Report Steel/Sem.10/R.4 (34 pages), Steel Committee of the Economie Commission for Europe (ECE) of the United Nations Organization (UNO), Geneva, Switzerland, January 1984.

Schouten, J.C., Metaal&Techniek, 29 (6), 40-42, 1984. Schouten, J.C., Metaal&Techniek. 29 (7), 28-31, 1984.

15. Grappige stellingen bestaan niet: als grappig bedoelde stellingen zijn meestal flauw, ver gezocht, vervelend of irritant en bovendien vaak stilistisch onnodig ingewikkeld.

16. De 'corrosiebeschermingsfactor' en de 'corrosie conditie' vormen bruikbare grootheden bij de technisch-economische analyse van corrosiekosten en mogelijke besparingen [1], zowel voor afzonderlijke bedrijven als voor totale bedrijfstakken [2].

[1] Schouten, J.C. and Gellings, P.J., Br. Corros. J., 19 (4), 159-164, 1984. [2] Schouten, J.C. and Gellings, P.J., J. Agr. Engng. Res.. 36, 217-231, 1987.

17. Het niveau, de kwaliteit en de werkwijze van de Nederlandse volksvertegen­ woordiging moeten worden verbeterd door:

a. uitbreiding van het aantal leden der Tweede Kamer van 150 naar 250; b. afschaffing van de Eerste Kamer;

c. het wijzigen van het parlementslidmaatschap van een 'volledige' in een 'deeltijd' functie: volksvertegenwoordigers moeten naast de functie van parlementslid andere politieke en/of maatschappelijke betrekkingen vervullen; bijvoorbeeld in het bedrijfsleven, de vakbond, het openbaar bestuur, de financiële wereld, het onderwijs, het wetenschappelijk onder­ zoek, de gezondheidszorg, de land- en tuinbouw enz.

d. een aanzienlijke vergroting van de bereidheid van oud-bewindspersonen om na het minister- of staatssecretarisschap voor een periode van ten minste vier jaren deel uit te maken van het parlement.

Extended Ep3teln - Helsenberg P r i n c i p l e .

In a R&D o r b i t , only two of the e x i s t i n g t h r e e parameters can be defined s i m u l t a n e o u s l y . These parameters a r e : t a s k , time and r e s o u r c e s .

1. If one knows what the t a s k i s , and t h e r e i s a time l i m i t allowed for the completion of the t a s k , then one cannot guess how much i t w i l l c o s t .

2. If the time and r e s o u r c e s a r e c l e a r l y defined, then i t i s impossible t o know what part of the R&D task w i l l be performed. 3 . I f one i s given a c l e a r l y defined R&D goal

and a d e f i n i t e amount of money which has been c a l c u l a t e d t o be necessary for the completion of the t a s k , one cannot p r e d i c t if and when the goal w i l l be reached.

I f one i s lucky enough and can a c c u r a t e l y d e f i n e a l l t h r e e p a r a m e t e r s , then what one d e a l s with i s not in the realm of R&D.

( i n : Arthur Bloch, Murphy's Law Complete - a l l the reasons why e v e r y t h i n g goes wrong, Methuen London L t d . , Great B r i t a i n , 1985.)

ACKNOWLEDGEMENTS

This thesis is a part of the research program on the 'egenerative aesulfurization during the fluidized bed combustion of coal as it is carried out at the Department of Chemical Engineering and Chemistry of the Delft University of Technology.

I would like to acknowledge gratefully:

* the financial support given by the Commission of the European Communities (Contract no. EN3F-001 t-NL (GDF)), the Netherlands Research Association 'Stichting voor de Technische Wetenschappen STW' (research project no. 700-3Ü9-173), the Netherlands Management Office for Energy Research PEO

(Contract no. 20.35-016.30) and Shell Internationale Petroleum Maatschappij B.V.;

* the succesful cooperation with the 'Institut National des Industries Extractives INIEX' in Liege (Belgium) in using different fluidized bed facilities.

Furthermore I am grateful to all persons who have contributed to the realization of this thesis by their scientific, technical or administrative support.

Because the mentioning of many names provides the possibility of forgetting some, I will only mention three: I want to thank Cor van den Bleek for his stimulating support and encouragement; further I want to thank Gerrit Hakvoort for his helpful suggestions; finally I am grateful to Henri Nasson for his enthousiastic cooperation.

CONTENTS

SUMMARY / SAMENVATTING

CHAPTER 1: INTRODUCTION and SCOPE of THESIS

1.1 The D.U.T. research on FBC 1 1.2 Overview of the research work: list of publications 1

1.3 Overview of the thesis

CHAPTER 2: SULFUR RETENTION during FLUIDIZED BED COMBUSTION of COAL 9

2.1 Introduction 9 2.2 J.C. Schouten and C M . van den Bleek

Prediction and Optimization of Sulfur Retention In the

Fluidized Bed Combustion of Coal: the D.U.T. SURE model 10 (submitted for publication in Chem. Eng. Sci., 1987)

2.3 J.C Schouten and C M . van den Bleek

The Influence of Oxygen-Stolchiometry on Desulfurization

during FBC: a Simple SURE Modeling Approach 63 (submitted for presentation at the 10th Int. Symp. on

Chemical Reaction Engineering, ISCRE 10, 1988)

2.« Comments 78

CHAPTER 3: PARTICLE MOTION in SLUGGING GAS FLUIDIZED BEDS 91

3.1 Introduction 91 3.2 J.C. Schouten, P.J.M. Valkenburg and C M . van den Bleek

Segregation in a Slugging FBC Large Particle System 92 (accepted for publication in Powder Technol., 1987)

3.3 J.C. Schouten, H.A. Masson and C M . van den Bleek

The Motion of Particles In a Slugging Gas Fluidized Bed 120 (submitted for publication in Powder Technol., 1987)

3." Comments 155

NOTATION 168 REFERENCES 172

SUMMARY

A key feature of fluidized bed coal combustion (FBC) is the possibility of the in-situ removal of the sulfur, by natural or synthetic sorbents, which is released from the burning coal. However, the consumption of the sorbent should be maintained at a reasonably low level in order to reduce the solid waste disposal as well as to achieve an acceptable cost of energy relative to conventional technologies. These considerations, together with others (like the interaction between S 0X- and NOx-removal), make clear that much research on its environmental performance is still needed before FBC 'comes of age'.

The research described in this thesis is predominantly focussed on two subjects, which are both related to the removal of sulfur during F3C:

1. the modeling of sulfur retention, and

2. the motion of particles in slugging gas fluidized beds. 1. the modeling of sulfur retention (chapter 2 ) .

Starting from an evaluation of retention models given in literature a simple analytical SUlfur REtention (SURE) model is described which can be used for the prediction and optimization of sulfur retention during FBC.

The SURE model is based on a simple approach to the sulfation kinetics: the sorbent sulfation rate is first order in the SO, (or S03) concentration and first order in the reactive sorbent surface area. Two-phase plug flow as well as one-phase Ideally mixed gas flow are considered.

The decrease of the desulfurizatlon efficiency during staged combustion as well as the temperature-dependency of the sulfure capture process can theoretically be explained by the formation of S 03 as a gaseous intermediate reactant in the sorbent sulfation reaction.

The model provides analytical equations for the dimensionless sulfur outlet concentration in case of unsteady state sulfation and for the molar calcium to sulfur ratio in the reactor feed during steady state operation; the latter being a function of fluid bed reactor conditions, coal and sorbent properties, and the stoichiometric in-bed air ratio.

The SURE model shows good agreement with experimental data obtained from the literature. It is argued that the sorbent residence time is an important system parameter, which importance is generally neglected during sulfur capture experimentations in pilot-plants.

Finally, the retention Index is introduced which is a useful parameter for optimization of the retention performance of a combustor. The index facilitates the choice of an appropriate action in the minimization of the calcium to sulfur feed ratio (e.g. change of particle size or superficial gas velocity).

2. the motion of particles In slugging gas fluidized beds (chapter 3 ) . The particles applied in FBC systems are generally B- or D-type of powders and differ significantly in size and density (coal, ash, sorfcent). Their fluidization behaviour Is strongly influenced by the presence of closely packed internals as heat exchanger tubes. Consequently, the fluidization is mostly of the slugging type and shows remarkable agreement with the mode of fluidization ir. relatively small diameter fluid beds.

Therefore slugging, particle mixing and segregation are investigated In

small dimeter beds C.l to 15 cm ID) with two different ntal techniques: abrupt defluidization and a radio-active tracer technique.

The slugging fcfd.3 are characterized by dimensionless slug parsreeters which are obtained from visual bed height measurements. It is observed that • --.csec slugging is likely to occur at Hm«./t-rat ios of tr.or? than ü. The distance between two su Ive pas slugs is found to b< Ir • pendent of ras velocity, while the average length of solids slugs equals about 80' to more than 901 of the bed height at minimum fluidization.

The extent of segregation increases with a larger H^j./0-ratio, a smaller bed diameter and a decreasing gas velocity. However, at a gas velocity of -•:-• times ■:■ Izatlor velocity still a significant degree of segregation is measured. Based on experimentations it is further suggested that segregation becor.es nil beyond a critical value of the bed diameter (about 23 cr, at a Hmf/C-ratlo of 2 and an excess gas velocity of 0.3 m / s ) .

Finally, dynamic radio-active tracer experiments are rreserted, which clearly show that a slugging bed consists of two regions: above and below the

-' -' r, of the ted height at rir-imum fluidization. The Influence is shown of the axial bed position, the gas velocity and the Hmf/D-ratio on the particle velocity, on the probability of changes in the direction of the particle's

lacement and en its presence probability.

The observed decrease of the presence probability along the bed height In a single-component bed is defined as 'solids dilution' and is explained by a balance of the drag force and gravity force acting on the particle.

Based or. these observations a general model for the motion of solids in slugging gas fluidized beds is introduced. The mechanism of solids motion Is based on convective and dispersive solids flows which differ in magnitude in the upper and lower part of the bed.

A comparison between the experimentally observed normalized presence probability of the tracer and the fitted solids concentration according to the model shows a good agreement which Is further applied for a diagnostic evaluation of solids motion in slugging gas fluidized beds.

SAMENVATTING

Een wezenlijke eigenschap van wervelbedverbranding van steenkool (WBV) vormt de mogelijkheid om de zwavel die vrijkomt bij de koolverbranding in-situ te vangen m.b.v. een natuurlijk of synthetisch sorbent materiaal. Het verbruik van dit sorbent moet echter in een aanvaardbare omvang worden gerealiseerd om enerzijds de vast-afval problematiek te beperken en om anderzijds een acceptabel niveau van energiekosten te verkrijgen In vergelijking met conventionele energietechnologieën. Deze overwegingen tezamen met anderen, zoals de Interactie tussen SO - and HO -verwijdering, maken duidelijk dat nog veel onderzoek ten aanzien van de milieuaspecten nodig is alvorens WBV als een 'volwassen' technologie kan worden aangemerkt.

Het onderzoek beschreven in dit proefschrift is voornamelijk gericht op twee onderwerpen, welke belde betrekking hebben op de verwijdering van zwavel tijdens wervelbedverbranding van steenkool:

1. de modellering van zwavelretentle, en

2. de beweging van deeltjes in sluggende, gas-gefluTdiseerde bedden. 1. de modellering van zwavelretentle (hoofdstuk 2 ) .

Uitgaande van een evaluatie van retentiemodellen in de literatuur, wordt een eenvoudig, analytisch model beschreven dat kan worden gebruikt voor de voorspelling en optimalisatie van zwavelretentle tijdens WBV. Dit model Is gebaseerd op een eenvoudige benadering van de sulfatatiekinetiek: de sorbent sulfatatiesnelheid is eerste orde in de S 02 (of S03) concentratie en eerste orde in het reactieve sorbent oppervlak. Zowel twee-fasen propstroming als de èèn-fase ideaal geroerde tank worden beschouwd.

De afname van de ontzwavelingsefficiency tijdens getrapte verbranding en de temperatuurafhankelijkheid van het zwavelvangstproces kunnen theoretisch worden verklaard met behulp van de vorming van S 03 als gasvormig tussen-produkt in de sorbent-sulfatatiereactie.

Het model biedt analytische vergelijkingen voor de dimensieloze zwavel-concentratie tijdens niet-stationaire sulfatatie en voor de molaire calcium-zwavel verhouding in de reactorvoeding tijdens stationaire operatie; de laatste is een functie van reactor condities, van kool- en sorbent-eigenschappen en van de stoichiometrische luchtverhouding in het bed.

Het model toont goede overeenstemming met experimentele gegevens uit de literatuur. Aangetoond wordt dat de sorbentverblijftijd een belangrijke systeemgrootheid is, welke In het algemeen wordt veronachtzaamd tijdens zwavelvangstexperimenten in (kleinschalige) proefopstellingen.

Ten slotte wordt de retentie-index geïntroduceerd, welke een bruikbare grootheid vormt voor de optimalisatie van de mate van retentie in een verbrandingsinstallatie. De index vergemakkelijkt de keuze van een geschikte aktie m.b.t. de minimalisatie van de molaire calcium-zwavel verhouding (bijv. verandering van de deeltjesgrootte of de superficiële gassnelheid).

2. deeltjesbeweging In sluggende gas-gefluldlseerde bedden (hoofdstuk 3 ) . De deeltjes die toegepast worden in WBV-systemen zijn doorgaans B- of D-type poeders en verschillen aanzienlijk in grootte en dichtheid (kool, as en sorbent). Hun fluTdisatiegedrag wordt sterk bernvloed door de aanwezigheid van de dichtgepakte warmtewisselaarpijpen in het bed (schijnbare beddiameter van ongeveer 10 c m ) . Dientengevolge wordt de fluldisatie meestal gekarakteri­ seerd '-oor 'slugging' en toont opvallende gelijkenis met de wijze van

fluldi-:r. wervelbedden van relatief kleine Inwendige diameter.

Daarom zijn slugging, deeltjesmenging en segregatie in kleine bedden (3.5 tot 15 cm ID) onderzocht met behulp van twee verschillende experimentele

:eken: abrupte defluidisatie en een radio-aktieve tracer techniek. De sluggende bedden worden gekarakteriseerd door dimensieloze slugpara-meters, welke worden verkregen m.b.v. visuele bedhoogtemetingen. Getoond wordt dat 'square-nosed slugging' optreedt bij Hmf/D-verhoudingen groter dan ■i. De afstand tussen twee opeenvolgende gasslugs is onafhankelijk van de gassr.-rlr.eid, terwijl de gemiddelde lengte van de deeltjesslugs ongeveer 80*

IfS r.eer dan 90* var. de bedhoogte bij minimale fluldisatie bedraagt. De mate van segregatie neemt toe met een grotere H^/D-verhouding, een - . -r.ere beddiameter en een afnemende superficiële gassnelheid. Bij een gassnelheid van driemaal de minimale fluldisatiesnelheid wordt echter nog steeds een significant niveau van segregatie gemeten. Op grond van experimen­ ten wordt verondersteld dat segregatie nihil wordt boven een kritieke grootte van de beddiameter (ongeveer 23 cm bij een Hmf-/D-verhouding van 2 en bij een gassnelheid van 0.3 m/s boven de minimale fluidisatlesnelheid).

Vervolgens worden dynamische radio-aktieve tracer experimenten gepresen­ teerd, welke duidelijk aantonen dat een sluggend bed bestaat uit een tweetal delen: onder en boven de positie van de bedhoogte bij minimale fluldisatie. De invloed wordt getoond van de axiale positie in het bed, de superficiële gassnelheid en de H ../D-verhoudlng, op de deeltjessnelheid, op de waarschijn-1 ijkheid van een verandering in de richting van de deeltjesverplaatsing en op de aanwezigheidswaarschijnlijkheid van het tracerdeeltje.

De waargenomen afname van de aanwezigheidswaarschijnlijkheid langs de bedhoogte in een bed van een enkelvoudige component, wordt gedefinieerd als 'deeltjesverdunning' en wordt verklaard met behulp van een evenwicht tussen de 'meesleurkracht' en de zwaartekracht welke op een deeltje werken.

Gebaseerd op deze waarnemingen wordt een algemeen model voor de beweging van deeltjes in een sluggend, gas-gefluldiseerd bed geïntroduceerd. Het mechanisme van deeltjesbeweging is gebaseerd op convectleve en dispersieve vaste-stof stromen, welke verschillen in grootte in het bovenste en onderste deel van het bed.

Een vergelijking tussen de experimenteel waargenomen genormaliseerde aanwezigheidswaarschijnlijkheid van de tracer en de 'gefitte' deeltjes­ concentratie volgens het model tonen een goede overeenstemming, welke verder wordt gebruikt voor een evaluatie van deeltjesbewegingen in sluggende bedden.

CHAPTER 1: INTRODUCTION and SCOPE of THESIS

1.1 The D.U.T. research on FBC

The research on sulfur retention during the fluidized bed combustion of coal at the Department of Chemical Engineering and Chemistry at the Delft University of Technology (D.U.T.) started in March 1980. The program is presently managed by prof. drs. P.J. van den Berg and ir. C M . van den Bleek.

The research was carried out from January 1980 to March 198" by PhD student ir. I.W. Noordergraaf (graduated May 1985). Noordergraaf started the work on sulfur capture during fluidized bed combustion of coal with use of a synthetic sorbent material. Further he paid attention to the external mass transfer phenomena and the fluid dynamical behaviour of large particles.

This work was continued in January 1984 and in August 1981 respectively by the PhD students ir. J.C. Schouten (financially supported by the Netherlands Research Organization STW) and ir. P.J.M. Valkenburg. A new aspect in the research program was the study with use of thermal analytical techniques of the release of sulfur during the combustion of coal.

In April 1986 a research project on the regenerative sulfur retention during FBC was started being financially supported by the Commission of the European Communities and the Netherlands Management Office for Energy Research PEO. Two PhD students have been appointed for the synthesis and testing of sorbent materials and for the design of a sorbent-regenerator (ir. A.E. Duisterwinkel and ir. E.H.P. Wolff).

1.2 Overview of the research work: list of publications

The work of the present author has predominantly been focussed on three topics in the D.U.T. research program:

1. the release of sulfur during coal combustion,

2. the retention of sulfur during the fluidized bed combustion of coal, 3. the motion of particles In slugging gas fluidized beds.

A collection of 14 papers has been written on these respective topics of the D.U.T. research program. The author of this thesis is first author of 11 papers and co-author of 3 papers. Some of these papers have already been published in conference proceedings or journals, while others have been submitted or accepted for presentation at conferences or publication in journals:

journals -published: 2 papers -accepted for publication: i paper -submitted for publication: ^ papers

conferences -published in proceedings: 6 papers -submitted for presentation: 1 papers

First, a list of these publications is given together with a short abstract of the papers so as to provide a comprehensive overview of the • r's work. Further, Id the next section of this chapter, it is indicated what the general scope of the present thesis will be and which papers therefore have been included in this thesis.

List of publications:

J_. the release of sulfur during coal combustion

1.1 Schouten, J.C., Hakvoort, C , Valkenburg, P.J.M, and Van den Bleek, C M . , "An Approach with Use of EGA to the Mechanism of Sulfur Release during Coal Combustion", Proc. 10th Nordic Symp. and Joint Nordic-English Sytnp. on Thermal Analysis and Calcrimetry, Bergen, Norway, Augu3t 20-22, 1985; in: Thermochimica Acta, JW, 171-178, 1987.

In this paper a reaction scheme for the release of sulfur from pyrite (FeS2) in coal is presented. The decomposition of pyrite is a complicated

process composed of several overlapping reactions. The reaction scheme is verified by thermogravimetric experiments with combustion of pyrite, FeS, coal and coal with additions of pyrite. The results are in good qualitative agreenent with the proposed reaction scheme.

1.2 Schouten, J.C., Blommaert, F.Y., Hakvoort, C. and Van den Bleek, C.M., "A Thermal Analytical Study on the Release of Sulfur during Coal Combustion", Proc. iq°7 Int. Conf. on Coal Science, Maastricht, The Netherlands, October 26-30, 1987; in: Coal Science and Technology 11, J.A. Moulijn, K.A. Hater and H.A.G. Chermin (Eds.), 837-81*0, Elsevier. Amsterdam, 1987.

In this second paper thermal analytical experiments with relatively high heating rates (up to 70 °C/min) ire presented. X-ray diffraction is used to determine the composition of specific samples at the end of relevant sulfur release peaks. The reaction products of pyrite combustion are FeS. ., FeS and F e203. FeSO» is not detected as such, however it can occur as an

intermediate reaction product. Seperate coal devolatilizatlon and subsequent char combustion experiments are carried out with three coal types with

-d i f f e r e n t s u l f u r compositions in or-der to -determine the -d i s t r i b u t i o n of s u l f u r over v o l a t i l e s and c h a r . The main conclusion i s t h a t in case of t h r e e d i f f e r e n t coal types a l l organic s u l f u r i s r e l e a s e d during coal d e v o l a t i l i z a t i o n , while p y r i t i c s u l f u r i s predominantly r e l e a s e d during the combustion of the c h a r .

1.3 Schouten, J . C . , Ingwersen, M.J. and Van den Bleek, C.M., "The Release of Sulfur from a Batch Addition of Coal t o a Fluid Bed", Proc. 1987 I n t . Conf. on Coal S c i e n c e , M a a s t r i c h t , The N e t h e r l a n d s , October 26-30, 1987: i n : Coal Science and Technology 11, J.A. Moulijn, K.A. Nater and H.A.G. Chermin ( E d s . ) , 811-8114, E l s e v l e r , Amsterdam, 1987.

In the l a s t paper In t h i s s e r i e s batch-wise s u l f u r r e l e a s e experiments a r e presented which a r e c a r r i e d out In a 5 cm ID diameter l a b o r a t o r y - s c a l e f l u i d l z e d bed. In case of one coal type i t Is confirmed t h a t the o r g a n i c s u l f u r i s r e l e a s e d during coal d e v o l a t i l i z a t i o n , while p y r i t i c s u l f u r i s oxidized during char combustion. In case of another coal type a d i f f e r e n t type of behaviour was found.

2 . t h e r e t e n t i o n of s u l f u r d u r i n g t h e f l u i d l z e d bed combustion of coal

2.1 Schouten, J . C . , Singh, P . C . , Valkenburg, P.J.M, and Van den Bleek, C.M., "Sulfur Release and Capture in F l u i d i z e d Bed Coal Combustion: A Review", submitted for p u b l i c a t i o n In Chem. Eng. Res. Des., 1986.

In t h i s review paper some Important v a r i a b l e s t h a t determine s u l f u r r e l e a s e are discussed as coal d e v o l a t i l i z a t i o n , the p o s i t i o n of s u l f u r g e n e r a t i o n and the e f f e c t of bed v a r i a b l e s as temperature and excess a i r . The advantages of s y n t h e t i c , r e g e n e r a t i v e s o r b e n t s a r e discussed v i s . a . v i s the more commonly a v a i l a b l e n a t u r a l s o r b e n t s as limestone or d o l o m i t e . The e f f e c t s of a d d i t i v e s , h y d r a t i o n , NO - i n t e r a c t i o n and freeboard are summarized, t o g e t h e r with v a r i a b l e s a f f e c t i n g s u l f u r c a p t u r e as the (Ca/S) r a t i o , p a r t i c l e s i z e , temperature, gas v e l o c i t y , p r e s s u r e , e t c . Further the review provides a summary of 22 t y p i c a l experimental s t u d i e s on s u l f u r c a p t u r e during FBC as well as an overview of t y p i c a l s u l f u r r e t e n t i o n models i n l i t e r a t u r e .

2.2 Hakvoort, C , Van den Bleek, C M . , Schouten, J . C . and Valkenburg, P . J . M . , "TG Study of Sorbent M a t e r i a l s for D e s u l f u r i z a t i o n of Combustion Gases a t High Temperature", Proc. 10th Nordic Symp. and J o i n t Nordic-English Symp. on Thermal Analysis and Calorlmetry, Bergen, Norway, August 20-22, 1986; i n : Thermochimlca Acta, 111, 103-108, 1987.

A thermogravimetrie study is summarized on the s u l f a t l o n and regeneration of CaO-irapregnated a - A l20 , ( p e l l e t s ) which is a synthetic sorbent that is used i n the D.U.T. research program.

2.3 Schouten, J.C. and Van den Bleek, C.H., "A C r i t i c a l Remark on the Use of Semi-empirical Models for Desulfurlzatlon ir. Fluid Bed Coal Combustion", Shorter Communication submitted for publication in Chem, Eng. S c i . , 1986.

In t h i s paper some serious concerns are brought forward with respect to the correct application of simple (semi-empirical) s u l f a t l o n k i n e t i c equations i n sulfur retention models as given i n the l i t e r a t u r e . I t is i l l u s t r a t e d that two s u l f a t i o n models i n l i t e r a t u r e are fundamentally i n c o r r e c t , because the mass conservation laws are not c o r r e c t l y applied.

2.1 Schouten, J.C. and Van den Bleek, C.M., "The D.U.T. SURE model: A Simple Approach i n FBC Sulfur Retention Modeling", Proc. 9th I n t . Conf. on Fluldized Bed Combustion, volume 2, 7^9-761, Boston, USA, May 3-7, 1987. A simple mathematical sulfur r e t e n t i o n (SURE) model is described that can be used f o r the prediction and optimization of the r e t e n t i o n of s u l f u r during FBC. S t a r t i n g from an evaluation of retention models ir. l i t e r a t u r e t h i s model is based on a simple approach t o the s u l f a t i o n k i n e t i c s : the sorbent s u l f a t i o n rate is f i r s t order in the gaseous s u l f u r concentration as w e l l as in the reactive sorbent surface. The model provides a simple a n a l y t i c a l expression for the molar (Ca/S) r a t i o i n the reactor feed as a function of the required l e v e l of s u l f u r retention i n the bed, the maximum CaO u t i l i z a t i o n of the sorbent and a dimensionless model parameter: the retention parameter.

2.5 Schouten, J.C. and Van den Bleek, C.M., "Prediction and Optimization of Sulfur Retention i n the Fluldized Bed Combustion of Coal: the D.U.T. SURE model", submitted for publication in Chem. Eng. S c i . , 1987.

In t h i s paper a non-steady state modeling approach is added to the steady state approach as i s outlined i n the previous paper. Hereto r e l a t i o n s are derived f o r the gaseous s u l f u r o u t l e t concentration ( s u l f u r break­ through) as a function of the dimensionless breakthrough time i n the case of unsteady state combustor operation due to a batch addition of sorbent or a s u l f u r step input on a bed containing a f i x e d amount or sorbent. The CaO conversion and the s u l f u r o u t l e t concentration are compared with l i t e r a t u r e data from which the s u l f a t i o n k i n e t i c reaction rate parameter i s calculated.

2.6 Valkenburg, P.J.M., Singh, P.C., Schouten, J.C. and Van den Bleek, C.M., " S u l f a t i o n Studies i n a Fixed Bed Reactor on Synthetic Sorbents f o r Possible Use i n the Regenerative Sulfur Capture Process i n Fluidized Bed Combustion of Coal", Proc. 9th I n t . Congr. on Chemical Engineering, CHISA 1987, Prague, Czechoslovakia, August 30 - September H, 1987. A f i x e d bed sorption study is presented in which the SURE model approach as outlined in the previous papers i s used t o describe the non-steady state s u l f a t i o n of the synthetic sorbent (paper 2.2) i n a fixed bed reactor. The influence of oxygen on the r a t e of s u l f a t i o n i s explained by the formation of the intermediate reactant SO,. In order to f i t the data the model i s extended to describe the s u l f a t i o n at two d i f f e r e n t reactive surfaces, which have k i n e t i c r e a c t i v i t i e s towards S03 uptake which d i f f e r an order i n magnitude.

2.7 Schouten, J.C. and Van den Bleek, C.M., "The Influence of Oxygen-Stoiehiometry on Desulfurization during FBC: A Simple SURE Modeling Approach", submitted f o r presentation at the 10th I n t . Symp. on Chemical Reaction Engineering, ISCRE 10, Basle, Switzerland, August 29 -September 1, 1988.

I n the last paper of t h i s series the influence of oxygen-stoichiometry on the desulfurization e f f i c i e n c y i n a FBC combustor i s discussed. The e f f e c t of a decreasing s u l f u r removal at low oxygen concentrations (at low in-bed a i r r a t i o s ) i s explained by the formation of SO,. The steady state SURE model i s extended and now provides an a n a l y t i c a l equation for the (Ca/S) r a t i o which i s also a function of the coal properties and of the i n -bed stoichiometric a i r r a t i o .

3. the motion of p a r t i c l e s i n slugging gas f l u i d i z e d beds

3.1 Schouten, J.C., Valkenburg, P.J.M, and Van den Bleek, C.M., "Segregation i n a Slugging FBC Large P a r t i c l e System", accepted for publication i n Powder Technology, 1987.

Segregation and slugging experiments are presented which have been carried out at ambient conditions with a binary large p a r t i c l e system representing the ash-coal/sorbent mixture i n a f l u i d bed combustor. The slugging beds are characterized by dimenslonless slug parameters which are obtained from visual bed height measurments. The influence of the s u p e r f i c i a l gas v e l o c i t y , the bed diameter and the bed aspect r a t i o on the extent of segregation i s investigated. A simple segregation model is introduced which is based on a mechanism of segregation i n slugging gas f l u i d i z e d beds that is proposed being based on experimental observations.

3.2 Schouten, J.C., Masson, H.A. and Van den Bleek, C.M., "A Model for the Segregation of Large Particles in Slugging Gas Fluldlzed Beds", Proc. Int. Symp. on Multiphase Flows, ISMF 1987, volume 1, 237-2«3, Zhejiang University, Hangzhou, People's Republic of China, August 3-5, 1987.

In this paper a simple approach is given to the modeling of segregation in (round- and square-nosed) slugging beds. The segregation mechanism is based on a segregative upwards flow of particles In the particle slug, a convective downwards flow of particles in the gas slug, together with particle dispersion of particle exchange between the particle and gas slug. The dispersion coefficients calculated with the model as a function of the superficial gas velocity and of the bed diameter are well in agreement with

the equation derived by Thiel and Potter.

3.3 Schouten, J.C., Masson, H.A. and Van den Bleek, C.M., "The Motion of Particles in a Slugging Gas Fluldlzed Bed", submitted for publication in Powder Technology, 1987.

An experimental study is presented with the objective to investigate with a dynamic radioactive tracer technique the movement of sorbent particles In a one- and two-component 10 cm ID fluid bed. Information is obtained on the presence probability of the tracer particle as a function of the bed height, the frequency of passage of the tracer at a given level, the probability for a direction change, the magnitude of the up- and downward velocities of the tracer as a function of the bed height and on the random diffusion- (or dispersion-) like character of the particles trajectories. This information is applied to obtain a qualitative description of the mechanism of particle motion in these systems. Furthermore, a general model for the motion of particles in slugging fluldlzed beds is formulated based on this mechanistic description.

3.U Valkenburg, P.J.M., Schouten, J.C. and Van den Bleek, C.M., "The Non-Steady State Segregation of Particles in Gas Fluldlzed Beds", Proc. 5th Eng. Found. Conf. on cluidlzation, Elsinore, Denmark, May 18-23, 1986; in: Fluidization, (Eds. K. Sstergaard and A. Sdrensen), Engineering Foundation, New ïork, 193-200, 1986.

In the last paper of this series a modeling approach is given on the non-steady state segregation of particles. It is shown that a pulse-like input of slightly denser tracer particles results in a significant maximum in the particle distribution curve. This maximum can be described with a non-steady state model equation which is based on the segregation model as Introduced In paper 3.1.

-1.3 Overview of the thesis

A selection of four papers has been included i n the present thesis (see section 1.2: papers 2.5, 2.7, 3.1 and 3.3); t h i s selection consists predominantly of papers which have not been published so far and which provide a comprehensive and c h a r a c t e r i s t i c overview of the authors work on the modeling of sulfur retention during FBC and the motion of p a r t i c l e s i n gas f l u i d i z e d beds.

The selected four papers are presented i n two chapters. In chapter 2 predominantly the modeling of sulfur retention during the f l u i d i z e d bed combustion of coal w i l l be considered. In chapter 3 the results obtained on d i f f e r e n t aspects concerning the motion and hydrodynamics of p a r t i c l e s i n gas f l u i d i z e d beds w i l l be discussed. The main subjects considered are slugging, segregation and mixing.

Each respective chapter begins with an introduction i n which a short overview i s given of the subject and the papers which are presented i n that s p e c i f i c chapter.

Furthermore, in the l a s t section of each chapter some comments on a l l the papers (including those not printed i n t h i s thesis) are added, i n order to increase t h e i r coherence or t o supply some extra information so as to enlarge the general understanding. Sometimes a formula or expression i n a comment may be a l i t t l e d i f f i c u l t to understand without having read the o r i g i n a l paper; however, the comments have been w r i t t e n i n such a way that the general 'message' which i t contains can easily be understood.

The notation applied i n the respective papers and comments is explained i n one overall notation l i s t which i s given at the end of the t h e s i s . This i s also the case with the references mentioned i n the papers and comments which are gathered i n one o v e r a l l l i s t .

F i n a l l y , i t should be noticed that the present text of the papers, as they are inserted i n the respective chapters In t h i s t h e s i s , might d i f f e r sometimes a l i t t l e from the text of the o r i g i n a l papers which have been submitted for p u b l i c a t i o n . P r i n c i p a l l y , t h i s has been done i n order to increase the mutual s t y l e and form; consequently, no fundamental alterations have been included.

CHAPTER 2: SULFUR RETENTION during FLUIDIZED BED COMBUSTION of COAL

2.1 I n t r o d u c t i o n

This chapter gives an impression of the t h e o r e t i c a l work which has been c a r r i e d out by t h e author on t h e d e r i v a t i o n of a mathematical model for t h e d e s c r i p t i o n of s u l f u r r e t e n t i o n during t h e f l u i d i z e d bed combustion of c o a l . No experimental r e s u l t s w i l l be presented or discussed other than those a v a i l a b l e in the l i t e r a t u r e : t h e s e experimental data a r e predominantly applied t o v a l i d a t e t h e proposed r e t e n t i o n model.

In the f i r s t paper (chapter 2.2) a simple mathematical s u l f u r r e t e n t i o n (SURE) model i s described which can be used for the p r e d i c t i o n and o p t i m i z a t i o n of the r e t e n t i o n of s u l f u r d u r i n g FBC. S t a r t i n g from an e v a l u a t i o n of r e t e n t i o n models in l i t e r a t u r e t h i s model i s based on a simple approach t o t h e s u l f a t l o n k i n e t i c s : t h e s o r b e n t s u l f a t i o n r a t e i s f i r s t order in the gaseous s u l f u r c o n c e n t r a t i o n , and f i r s t order i n t h e r e a c t i v e s o r b e n t s u r f a c e . The model provides a simple a n a l y t i c a l expression for t h e molar (Ca/S) r a t i o in the r e a c t o r feed as a function of a. the r e q u i r e d l e v e l of s u l f u r r e t e n t i o n i n t h e bed, b . the maximum CaO u t i l i z a t i o n of the s o r b e n t , and c . a dimensionless model parameter: t h e r e t e n t i o n parameter.

Furthermore r e l a t i o n s a r e derived for t h e gaseous s u l f u r r e a c t o r o u t l e t c o n c e n t r a t i o n ( s u l f u r breakthrough) as a function of the dimensionless breakthrough time i n t h e case of unsteady s t a t e combustor o p e r a t i o n due t o a batch a d d i t i o n of s o r b e n t or a s u l f u r s t e p input on a bed c o n t a i n i n g a fixed amount of s o r b e n t . The CaO conversion and t h e s u l f u r o u t l e t c o n c e n t r a t i o n a r e compared with l i t e r a t u r e data from which t h e s u l f a t i o n k i n e t i c r e a c t i o n r a t e parameter i s c a l c u l a t e d .

In t h e second paper (chapter 2.3) t h e i n f l u e n c e of oxygen-stoichiometry on t h e d e s u l f u r i z a t i o n e f f i c i e n c y i s d i s c u s s e d . The e f f e c t of a decreasing s u l f u r removal a t low oxygen c o n c e n t r a t i o n s ( a t low in-bed a i r r a t i o s ) i s explained by t h e formation of SO, as a gaseous i n t e r m e d i a t e r e a c t a n t in the s u l f a t i o n r e a c t i o n . The steady s t a t e SURE model i s extended and now provides an a n a l y t i c a l equation for t h e (Ca/S) r a t i o which i s a l s o a function of the coal p r o p e r t i e s and of the ln-bed s t o i c h l o m e t r i c a i r r a t i o . A good q u a l i t a ­ t i v e agreement with l i t e r a t u r e data i s r e p o r t e d . Furthermore t h e Importance of t h e sorbent r e s i d e n c e time with r e s p e c t t o model v a l i d a t i o n i s d i s c u s s e d .

-2 . -2 PREDICTION and OPTIMIZATION of SULFUR RETENTION I n the FLUIDIZED BED COMBUSTION o f COAL:

t h e D.U.T. SURE model

s u b m i t t e d f o r p u b l i c a t i o n I n Chem. Eng. S c l . , 1987

SUMMARY 12 1 . INTRODUCTION 12

1.1 S u l f u r r e l e a s e and c a p t u r e d u r i n g FBC 12

1.2 S u l f u r r e t e n t i o n modeling 11

2 . THE SURE MODEL 19 2.1 S u l f a t i o n k i n e t i c term 19

2 . 2 Choice of f l u i d dynamical r e a c t o r model 19 2 . 3 Summary of SURE model assumptions 20

2 . 3 . 1 One-phase model 20 2 . 3 . 2 Two-phase model 21 2 . 3 . 3 Both models 21 3. SULFUR BREAKTHROUGH AT UNSTEADY STATE OPERATION 22

3.1 One-phase model 22 3.2 Two-phase model 21 1 . (Ca/S) RATIO AT UNSTEADY STATE OPERATION 25

1.1 One-phase model 25 1.2 Two-phase model 27 5 . COMPARISON OF MODEL CALCULATIONS WITH EXPERIMENTAL DATA 28

5.1 S u l f a t i o n k i n e t i c term 28 5.2 S u l f u r breakthrough a t unsteady s t a t e o p e r a t i o n 32

5 . 2 . 1 CaO conversion 32 5 . 2 . 2 S u l f u r o u t l e t c o n c e n t r a t i o n 33

5.3 S u l f u r r e t e n t i o n a t steady s t a t e o p e r a t i o n 31

6 . MODEL PARAMETER SENSITIVITY ANALYSIS 39 6.1 C a l c u l a t i o n of model parameters 39 6 . 2 Comparison o f one- and two-phase model 39

6 . 2 . 1 S u l f u r b r e a k t h r o u g h a t unsteady s t a t e o p e r a t i o n 39

6 . 2 . 2 (Ca/S) r a t i o a t steady s t a t e o p e r a t i o n 10

6.3 Level of s u l f u r r e t e n t i o n 11 6 . 3 . 1 I n f l u e n c e of maximum s o r b e n t conversion 11

6 . 3 - 2 I n f l u e n c e o f r e t e n t i o n parameter 15 6 . 1 I n f l u e n c e of r e a c t o r circumstances and sorbent p r o p e r t i e s

on s u l f u r r e t e n t i o n 16 6 . 1 . 1 Sorbent r e s i d e n c e time 17

6 . 1 . 2 Sorbent p a r t i c l e s i z e 18 6 . 1 . 3 S u p e r f i c i a l gas v e l o c i t y 19

7 . OPTIMIZATION OF THE (Ca/S) RATIO 50

7.1 R e t e n t i o n index RI 50

7.2 Choice of parameter change cpc 52

8 . CONCLUSIONS 5^ APPENDIX 1: THE MODEL PARAMETERS 55

1 . Maximum gas exchange r a t i o P„ 55

2. Two-phase gas flow parameter m 58

3. Retention parameter M 59 ■<. Maximum conversion a 60 5 . Magnitude of model parameters 60

APPENDIX 2 : MODEL LIMITATION DUE TO OXYGEN CONSUMPTION 61

-PREDICTION and OPTIMIZATION of SULFUR RETENTION i n t h e FLUIDIZED BED COMBUSTION of COAL:

t h e D.U.T. SURE model

J . C . S c h o u t e n and C M . van den B l e e k

SUMMARY

In t h i s paper t h e D.U.T. SU(lfur)RE(tention) model i s described which i s a simple mathematical model t h a t can be used for t h e p r e d i c t i o n and o p t i m i z a t i o n of t h e r e t e n t i o n of s u l f u r during t h e f l u i d i z e d bed combustion of c o a l . S t a r t i n g from an e v a l u a t i o n of r e t e n t i o n models given i n l i t e r a t u r e the SURE model i s based on a simple approach t o t h e s u l f a t l o n k i n e t i c s : t h e s o r b e n t s u l f a t i o n r a t e i s a. f i r s t order i n t h e gaseous s u l f u r component c o n c e n t r a t i o n , and b . f i r s t order in t h e r e a c t i v e s o r b e n t s u r f a c e . The f l u i d dynamical part of t h e SURE model i s based on t h e two-phase theory of f l u i d l z a t i o n with plug flow of ga3 In the bubble phase and an I d e a l l y mixed dense p h a s e . Also t h e c a s e of a high gas exchange between both phases i s considered r e s u l t i n g i n a one-phase i d e a l l y s t i r r e d tank model. In t h e case of steady s t a t e combustor o p e r a t i o n t h e one-phase SURE model provides a simple a n a l y t i c a l expression for the molar (Ca/S) r a t i o i n t h e r e a c t o r feed as a function of a . t h e r e q u i r e d l e v e l of s u l f u r r e t e n t i o n in the bed, b . t h e maximum CaO u t i l i z a t i o n of t h e s o r b e n t , and c . a dimenslonless model parameter: the r e t e n t i o n parameter. The one-phase SURE model i s compared with s u l f a t l o n experiments derived from t h e l i t e r a t u r e in o r d e r t o check I t s general v a l i d i t y . A parameter s e n s i t i v i t y a n a l y s i s i s given form which t h e r e l a t i v e importance of the r e s p e c t i v e (dimenslonless) model parameters I s deduced. F i n a l l y , t h e r e t e n t i o n index i s introduced which i s a parameter t h a t i s very useful for t h e judgement of the s u l f u r r e t e n t i o n performance of t h e combustor. I t can be used t o optimize both t h e f l u i d bed o p e r a t i n g c o n d i t i o n s and t h e sorbent p r o p e r t i e s in order t o o b t a i n t h e minimum p o s s i b l e (Ca/S) r a t i o a t a s p e c i f i c r e q u i r e d l e v e l of s u l f u r r e t e n t i o n .

1 . INTRODUCTION

1.1 Sulfur r e l e a s e and c a p t u r e during FBC

F l u i d i z e d beds are used for many d i f f e r e n t chemical e n g i n e e r i n g a p p l i c a t i o n s . One of t h e s e i s t h e production of energy from c o a l , although t h i s technique I s s t i l l under development and many problems concerning t h e fundamentals of f l u i d l z a t i o n , combustion, gas t r a n s f e r , heat exchange e t c . have t o be solved yet (LaNauze, 1985).

One of the advantages of the f l u i d i z e d bed coal combustion technique i s the p o s s i b i l i t y of t h e i n - s i t u removal of t h e s u l f u r r e l e a s e d from t h e coal by the a d d i t i o n of a n a t u r a l sorbent m a t e r i a l (limestone or dolomite) or a s y n t h e t i c s o r b e n t . Much r e s e a r c h has been c a r r i e d out i n t h i s f i e l d , experimentally as well as t h e o r e t i c a l l y : a comprehensive l i t e r a t u r e review on s u l f u r r e l e a s e and c a p t u r e i n t h e f l u i d i z e d bed combustion of coal i s provided by Schouten e t a l . (1986).

They concluded that in general 25Ï to 60? of the sulfur in coal is

released with the volatiles; the absolute percentage Increases with

increasing temperature and is probably not dependent on total volatile

yield. In general more than 90? of the sulfur in volatiles is present as

H2S. The major p o r t i o n of the v o l a t i l e s u l f u r i s r e l e a s e d near the coal feed p o i n t , while the remaining s u l f u r i s r e l e a s e d during subsequent char combustion in t h e bed.

I t i s c l e a r t h a t under o x i d i z i n g circumstances in t h e bed the s u l f u r i s predominantly r e l e a s e d i n the form of s u l f u r d i o x i d e which o r i g i n a t e s from the combustion of the o r g a n i c and p y r i t i c s u l f u r (FeS2) in the c o a l . In general i t can be assumed t h a t t h e r a t e of r e l e a s e of s u l f u r from c o a l , a t high combustion r a t e s as p r a c t i s e d in f l u i d bed combustion, i s l i n e a r l y dependent on t h e r a t e of coal combustion (Schouten e t a l . , 1986; Schouten e t a l . , 1987c). A r e a c t i o n scheme for t h e r e l e a s e of s u l f u r from p y r i t e i n coal i s given by Schouten e t a l . (1987c).

In a reducing atmosphere i t i s p o s s i b l e t h a t a l s o hydrogen s u l f i d e and COS a r e e m i t t e d , where H2S for example o r i g i n a t e s from the r e a c t i o n of hydrogen with the p y r i t e i n c o a l :

FeS2 + H2 -> H2S + FeS

The formation of t h e s e reduced s u l f u r s p e c i e s i s of importance i n t h e m u l t i - s t a g e s u b - s t o i c h i o m e t r i c f l u i d bed coal f i r i n g which may be necessary i n order t o reduce the NO -emission s i g n i f i c a n t l y (as for example in t h e l i m e s t o n e i n j e c t i o n m u l t i s t a g e burner LIMB of the U.S. Environmental P r o t e c t i o n Agency; Borgwardt e t a l . , 198*1).

In general lime o f f e r s the p o s s i b i l i t y of t h e c a p t u r e of s u l f u r in t h e case of S02 as well as of t h e reduced s u l f u r s p e c i e s H2S and COS (Simons and Rawlins, 1980; Borgwardt e t a l . , 1984) a c c o r d i n g t o the following r e a c t i o n s :

CaO ♦ S 02 + ' /2 02 -» CaSO,

CaO ♦ H2S -» CaS + H20

CaO ♦ COS ■» CaS + CO,

I t i s important t o n o t i c e t h a t in t h e case of hydrogen s u l f i d e c a p t u r e the calcium u t i l i z a t i o n can reach 100$ when i n i t i a l p a r t i c l e p o r o s i t i e s a r e g r e a t e r than k2% (Simons and Garman, 1986), while t h e maximum conversion in the once-through use of l i m e - s o r b e n t in t h e case of S02 c a p t u r e i s g e n e r a l l y smaller than 15% due t o t h e d i f f e r e n c e i n t h e molar volumes of the r e s p e c t i v e r e a c t i o n p r o d u c t s , which i n f l u e n c e s the occurence of pore plugging. F u r t h e r Simons and Rawlins (1980) concluded t h a t any procedure

-which u t i l i z e s limestone removal of SO, i s p o t e n t i a l l y capable of removing H,S a t approximately t h e same r a t e .

So lime-based s o r b e n t s can be used in o x i d i z i n g as well as reducing a t m o s f e r e s . However, t h e c a p t u r e of H2S and COS instead of S02 may have important i m p l i c a t i o n s for t h e r e s e a r c h on t h e r e g e n e r a t i v e removal of s u l f u r by n a t u r a l and s y n t h e t i c sorbent m a t e r i a l s as p r e s e n t l y c a r r i e d out In The Netherlands a t the Delft U n i v e r s i t y of Technology and Twente U n i v e r s i t y (Kamphuis e t a l . , 1987).

So i t i s concluded t h a t the type of s u l f a t i o n process g r e a t l y i n f l u e n c e s the optimum use of t h e s o r b e n t m a t e r i a l . However, a l s o other phenomena and parameters are of great importance for t h e p r e d i c t i o n and o p t i m i z a t i o n of s u l f u r r e t e n t i o n in f l u i d bed coal combustion. Here we can think of for example r e a c t o r o p e r a t i n g conditons l i k e the (Ca/S) r a t i o or t h e s u p e r f i c i a l gas v e l o c i t y , but a l s o sorbent p r o p e r t i e s as pore volume and p a r t i c l e s i z e (Schouten e t a l . , 1986). In t h i s paper i t i s t r i e d t o e s t a b l i s h which parameters and phenomena a r e important and i n what way they can e v e n t u a l l y be influenced t o obtain an optimum high l e v e l of s u l f u r r e t e n t i o n i n a f l u i d bed combustor.

1.2 Sulfur r e t e n t i o n modeling

One useful way t o determine t h e most important parameters t h a t i n f l u e n c e t h e s u l f a t i o n performance of a f l u i d bed coal combustor i s t o o b t a i n a mathematical r e a c t o r model which d e s c r i b e s the main chemical and physical

phenomena. A d e t a i l e d a n a l y s i s of model c a l c u l a t i o n r e s u l t s , t o g e t h e r with a model parameter s e n s i t i v i t y a n a l y s i s , w i l l in many cases give useful i n s i g h t i n t h e s u l f a t i o n p r o c e s s .

At l e a s t «0 o v e r a l l r i u i d i z e d bed coal combustion system models have been published i n l i t e r a t u r e . The majority of t h e s e models has been reviewed by Olofsson (1980), Park e t a l . (1980), LaNauze and Jung (1982) and LaNauze (1985). These reviews a r e based on a summary of t h e basic model assumptions. However, t h e s e summaries a r e e s s e n t i a l l y point by point comparisons of the various models, r a t h e r than c r i t i c a l assessments of t h e models's p r e d i c t i v e c a p a b i l i t i e s or a c c e s s a b i l i t y ( P r e t o , 1985).

LaNauze (1985) concludes t h a t in general t h e models meet only with l i m i t e d succes in t h e i r a b i l i t y t o p r e d i c t combustion performance, carbon loading or gas c o n c e n t r a t i o n and temperature p r o f i l e s . The accuracy with which some c r i t i c a l parameters are known or can be estimated i s not g r e a t , which allows r e a s o n a b l e v a r i a t i o n in p r e d i c t i o n s by a d j u s t i n g parameters such as the cross-flow f a c t o r , bubble s i z e or chemical r a t e . Preto (1985) adds t o t h i s t h a t the a v a i l a b l e models a r e In some cases overcomplicated,

with the inclusion of i r r e l e v a n t minutae, and i n other cases too rudimentary, with the exclusion of important f a c t o r s .

I n most cases the v a l i d i t y of these overall system models can be questioned as for example i s demonstrated for the s u l f u r retention prediction of the MIT system model (Molayem et a l . , 1982). However, also comprehensive models have been developed ( e . g . Wells et a l . , 1982), which are reasonably accurate, but these are also quite complicated and require s i g n i f i c a n t computing f a c i l i t i e s .

I t i s remarkable that only a few models as reviewed by LaNauze (1985) incorporate sulfur release as well as sulfur capture. Therefore eight t y p i c a l f l u i d bed retention models have been l i s t e d i n Table 1 which showed a reasonable to very good agreement between the calculated prediction of the s u l f u r capture e f f i c i e n c y and experimental combustor data (Ho et a l . (1986) presented only model calculations based on experimental data). I t i s remarkable that the modeling of the s u i f a t l o n process i n only two of the models summarized i n Table 1 is based on a fundamental s t r u c t u r a l gas-solid reaction model (see Ramachandran and Doraiswamy (1982) for a comprehensive review of these models). Many workers have used these type of models f o r the modeling of the s u l f u r retention reaction of calcium oxide and s u l f u r dioxide In other types of reactors (as d i f f e r e n t i a l reactors or thermobalances).

The common gas-solid reaction models which have been i n use so f a r are the k i n e t i c model (for example Borgwardt (1970) and Borgwardt and Harvey (1972)), the grain model (Ishida and Wen, 1971; Wen and I s h i d a , 1973; Ranade and Harrison, 1979), the pore model (Ramachandran and Smith, 1977; Georgakis et a l . , 1977; Chrostowskl and Georgakis, 1978; Georgakis et a l . , 1979; Dogu, 1981; Bhatia and Perlmutter, 1981; Chrlstman and Edgar, 1983; Bhatia, 1985; Simons and Carman, 1986) and the volume reaction model (Fan et a l . , 1977; 1978; 1981)).

One major disadvantage of the application of s t r u c t u r a l gas s o l i d reaction models t o overall f l u i d bed combustion system models Is mentioned by Fieldes (1979), Lee and Georgakis (1981), Fee et a l . (1983) and Dennis and Fieldes (1986). They argued that detailed reaction models that take i n t o account changes of s t r u c t u r a l s o l i d properties are highly non-linear. Therefore i t i s very d i f f i c u l t to Incorporate them easily i n t o f l u i d dynamical models of a f l u i d i z e d bed combustor without resort to excessively long computer calculations for each size of stone i n the feed. This w i l l espcially hold for the mathematical modeling of the unsteady state s u l f u r release and absorption of s u l f u r from a batch addition of coal and/or sorbent m a t e r i a l . Such a rigorous analysis of the S02 or H2S/COS s u i f a t l o n

-Table 1: Overview of t y p i c a l f l u i d bed s u l f u r r e t e n t i o n models, which show good agreement with experimental combustor d a t a ;

(see for d e s c r i p t i o n Table 2 ) .

A u t h o r ( s ) B e t h e l l e t a l . (1973)

Chen and Saxena (1977)

Raj an e t a l . (1978)

Raj an and Wen ' l v - 0 )

Lee e t a l . (1980) Lee and Georgakis (1981) Zheng e t a l . (1982) Fee e t a l . (1983) Fee e t a l . (1984) Ho e t a l . (1986) A 1 3 2a 2a 2b 2b 2c 2e 2c B ■ 1 2 2 3 ? 3 3 1 C D E F G H 1/yes 1*2 3a yes/2 no 2 1 1/yes 2/yes 2/yes 1/no 1/no 1/no 1/no 1/no ♦2 3b y e s / 2 no 2 la yes no 2 ►3 l a 2*3b 2*3c 2*3b 2+3b lb yes -.'. r.c, nc

nc

( t h i s work)

2c 1/no 3c"J _no

Table 2: D e s c r i p t i o n of t y p i c a l f l u i d bed s u l f u r r e t e n t i o n models as summarized i n Table 1.

A. Fluid bed model. 1) One phase model.

2a) Two phase bubbling bed model, clouded bubbles; bubble phase and emulsion phase; bubble s i z e dependent on bed h e i g h t ; minimum f l u l d i z a t l o n c o n d i t i o n s in the emulsion phase; gas exchange between bubble and emulsion phase i s a x i a l l y d i s t r i b u t e d (compartments in s e r i e s model).

2b) Two phase bubbling bed model; bubble phase and emulsion phase; average bubble s i z e ; minimum f l u l d i z a t l o n c o n d i t i o n s i n t h e emulsion phase; average gas exchange c o e f f i c i e n t .

Table 2 continued . . .

2c) Two phase bubbling bed model; bubble phase and emulsion phase; average bubble s i z e ; minimum f l u i d i z a t i o n c o n d i t i o n s i n t h e emulsion phase; gas exchange between phases i s uniform or non-uniform a c c o r d i n g t o f a s t and slow bubble regimes.

3) Three phase bubbling bed model: bubble, cloud-wake and emulsion phase; bubble s i z e i s dependent on bed h e i g h t ; minimum f l u i d i z a t i o n c o n d i t i o n s in t h e emulsion phase; gas exchange between phases i s based on average bubble volume.

B. Gas flow p a t t e r n .

1) Plug flow of gas in a l l phases. 2) Mixed flow in a l l p h a s e s .

3) Plug flow in bubble phase; emulsion phase well mixed. C. S o l l d 3 flow.

1) S o l i d s i d e a l l y mixed ( i n emulsion p h a s e ) . 2) S o l i d s well mixed in a number of compartments. 3) S o l i d s diameter d i s t r i b u t i o n i s considered: yes / no. D. Sulfur r e l e a s e .

1) Sulfur r e l e a s e r a t e i s p r o p o r t i o n a l t o coal ( c h a r ) combustion r a t e (or coal feed r a t e ) .

2) Sulfur r e l e a s e i s not uniform over t h e bed h e i g h t .

3) Sulfur r e l e a s e during d e v o l a t l l i z a t i o n i s considered and i s taken temperature dependent.

E. Sulfur c a p t u r e .

1a) Grain model of Wen and I s h i d a (1973) i s used t o c a l c u l a t e limestone r e a c t i v i t y .

1b) Grain model of Hartman and Coughlir. (1976) i s used. 2) Semi-empirical s u l f u r c a p t u r e k i n e t i c equation i s used. 3) K i n e t i c r a t e data obtained from:

a) fixed bed l a b o r a t o r y experiments; b) thermogravimetric a n a l y s i s ; c) f l u i d bed batch experiments.

1) S u l f a t i o n r a t e i s l i n e a r p r o p o r t i o n a l t o r e a c t i v e sorbent s u r f a c e . F. E l u t r l a t l o n .

1) E l u t r i a t i o n i s c o n s i d e r e d : yes / no. 2) R e c i r c u l a t i o n of e l u t r i a t e d p a r t i c l e s . G. Freeboard.

1) Freeboard s u l f u r r e t e n t i o n i s considered: yes / no. H. Model complexity.

1) Simple a n a l y t i c a l expressions a r e a p p l i e d . 2) More complex (numerical) c a l c u l a t i o n s a r e needed.

reaction results i n mathematical expressions that are too complicated for p r a c t i c a l engineering applications.

Further Fieldes (1979) concluded that using a computer to solve equations describing p a r t i c l e s u l f a t i o n also divorces the user sometimes Trom physical r e a l i t y , as he i l l u s t r a t e d c l e a r l y with two mistakes i n the application of s o l i d reaction models i n l i t e r a t u r e . I n one case he showed that s t i l l reasonable f i t s of s o l i d conversion against time were derived i n the work of Hartman and Couglin (1976) with an Incorrect value of the e f f e c t i v e surface rate constant forcing the adjusted e f f e c t i v e product d i f f u s i o n c o e f f i c i e n t to a rather low value ( t h i s was also noticed by Simons and Rawlins (1980) who suggested that the d i f f u s i o n c o e f f i c i e n t was a factor 20 to low).

Consequently, another important disadvantage of the s t r u c t u r a l gas-solid reaction models is that they require thorough knowledge of p a r t i c l e properties as pore size d i s t r i b u t i o n and grain diameter that are strongly stone dependent, as well as physical constants as the s o l i d d i f f u s i o n c o e f f i c i e n t which are extremely d i f f i c u l t t o determine experimentally. So Tar the s o l i d - s t a t e d i f f u s i o n c o e f f i c i e n t of SO, through a nonporcus layer of CaS0„ was obtained by adjusting i t i n a pore or grain model. A comprehensive summary of the d i f f u s l v l t y estimates as obtained by d i f f e r e n t authors i s given by Marsh and Ulrichson (1985).

However, i t must be noticed that the disadvantages as mentioned above of course do not detract from the conceptual contribution of the fundamental gas-solid reaction models t o the basic understanding of the influence of the s o l i d s s t r u c t u r e (grains and pores) on the overall reaction r a t e .

To avoid the possible extensive computation as well as the problem of experimental determination of physical constants or s o l i d properties, a more recent trend i n the f l u i d bed modeling of s u l f u r r e t e n t i o n is based on the experimental observation that the decay i n the o v e r a l l solids conversion can easily be f i t t e d by an exponential, reciprocal or other type of empirical r e l a t i o n (see Table 1). These models provide simple a n a l y t i c a l expressions f o r the s u l f u r retention as a function of the molar (Ca/S) r a t i o that can easily be applied to real combustors ( f o r example Lee et a l . , 1980; Lee and Georgakls, 1981; Zheng et a l . , 1982; Fee et a l . , 1983; Fee et a l . , 1981; Noordergraaf et a l . , 1985; Dennis and Fieldes, 1986). The sorbent parameters in these models, which describe the s u l f a t i o n k i n e t i c s , are obtained either from thermogravlmetrlc analysis data ( e . g . Fee et a l . , 1983; 1981) or from batchwise s u l f a t i o n experiments i n f l u i d beds (Zheng et a l . , 1982; Noordergraaf et a l . , 1985; Dennis and Fieldes, 1986). However, seme c r i t i c a l